DE4103808C2 - Device for non-destructive crack and dimension testing of thin-walled tubular elements - Google Patents

Description

The invention relates to a device for non-destructive crack and dimension testing of thin-walled, tubular elements.

For certain applications, the dimensional and structural integrity of certain critical components of the utmost importance to future failures safe to avoid. An example of one critical component is the liner tube of one Nuclear fuel rod that contains the fuel pellet column. With the open ends of the liner tube become plugs welded to the pellet column contained therein to include. The lining pipes must be under strict Standards of structural integrity are made, if you want to withstand the high internal pressures that over the long life of nuclear fuel rods be developed. That means cracks like  for example fractures, pores etc., in the pipe wall Cross section depending on its size, number and location a lining pipe can unsafe for one Use in a nuclear fuel rod.

Lining pipes also have strict dimensional standards fulfill. The inside diameter must be kept precisely so that the fuel pellets are introduced correctly can be. The same goes for the outside diameter, so that the fuel rods properly form fuel bundles can be put together. The wall thickness is also a rejection criterion because a thin wall cross-section, which is smaller than a minimum tolerance dimension, the Internal pressure resistance endangered.

Because of the critical nature of the lining pipes from Nuclear fuel rods it is necessary to use each one To inspect the lining non-destructively, specifically via its entire length both on their dimensional as well structural integrity before being in a Nuclear fuel rod can be used. It is currently usually using an ultrasonic test a converter that is operated in pulse echo mode, used to critical components regarding their To examine quality. The converter feels the component ab by moving the transducer and / or the component, while the transducer is periodically energized to emit a testing ultrasound energy pulse, and in the intervals between the pulses the Transducers the echoes that contain test information. The for Testing each component time is strong depending on the scanning speed. Of course the scanning speed may not be so great that the Converter "runs away" from the testing energy pulse so that it the echoes associated with each pulse are not adequate receives. To this limitation regarding the Compensating the scanning speed are numerous Transducers have been used to shorten the test time.  

This solution in turn increases the hardware cost of an ultrasonic test facility, since each transducer has one separate signal channel required, each its own Electronics needed to get the test information from the Pull out echo signals.  

US-A-4 475 399 describes a device for ultrasound Crack test for non-rotating tubular goods. she requires a large number of ultrasonic transducers and can only check the wall thickness and the presence of cracks.

Furthermore, US-A-3 828 609 describes a destruction Free ultrasonic test facility for long-term testing workpieces at high speed through a nested scanning system, the entire volume at different angles and several different, overlapping and nested scan patterns can be examined.

It is an object of the present invention to provide a device of the type mentioned at the beginning, with which one full quality assurance check by langge stretched thin-walled tubular elements to inexpensive and can be done efficiently.

The object is achieved by the features of Claim 1 solved.

Advantageous embodiments of the invention are in the Claimed claims.

The advantages which can be achieved with the invention are in particular special in that a high-speed digitali is obtained with which at all times a Determination of the location of cracks, a determination of Dimensions, a more flexible compensation of sound speed in water and a greatly reduced System setting can be achieved when using pipes different diameters can be checked. Doing so only a very simplified hardware is required, with one small number of ultrasonic transducers both dimensional as well as crack echoes determined.

The invention will now based on the description and drawing of Exemplary embodiments explained in more detail.

Fig. 1 is a schematic perspective view of an ultrasonic test station used in the device according to the invention for testing thin-walled tubular elements.

FIG. 2 is a schematic sectional view and shows the dimensions-testing ultrasound transducer and the fluid-coupling and temperature-compensating transducer used in the test station according to FIG. 1.

FIGS. 3 and 4 are schematic cross-sectional images and represent the Rißprüfungs ultrasonic transducer, which in the test station of FIG. 1 are used.

Fig. 5 is a block diagram of the ultrasonic testing device according to the invention.

FIG. 6 is a timing signal diagram and illustrates the operation of the device of FIG. 5.

FIG. 7 illustrates signal diagrams from the operation of the echo signal discriminators in FIG. 5.

In Fig. 1, an ultrasonic testing device is shown according to the invention, the X1 uses at least seven ultrasound transducers to X7, which are mounted at an inspection station 10 by suitable means (not shown) through which a passing axially or longitudinally to be tested metal tube 12, as indicated by arrow 16 . The tube and transducers are immersed in a bath (not shown) with a suitable liquid coupling agent, such as water, which serves to effectively couple the highly focused bundles of ultrasound energy testing pulses through the transducers be emitted. Since the tube moves both axially and in the direction of rotation, these test bundles follow individual helical scanning paths of great slope over the circumferential surface of the tube along its length. The pulse rate of the converter is sufficiently fast to ensure an intensive check of the entire pipe cross-section for both dimensional acceptance and for the presence of cracks. It will be appreciated that, instead of rotating the tube, the helical scan can be obtained by rotating the transducers about the tube axis as the tube is transported axially through the test station.

Of the ultrasonic transducers shown in FIG. 1, the transducers X1 and X2 are used to check the dimensions of the tube, which can be a nuclear fuel rod lining tube. Thus, as can be seen from FIG. 2, these two transducers are arranged on opposite sides of the tube 12 , their ultrasonic energy beams 18 being aligned diametrically. The transducers are separated by a known, fixed distance D. When transducers X1 and X2 are electrically pulsed to deliver ultrasound energy pulses, often referred to as main pulses, a series of echoes from each main pulse are received by the outer tube surface, the inner tube surface and a second echo is received by the inner tube surface . These three echoes, which are received by each transducer X1 and X2, are processed together with the distance D to determine the outer diameter, the inner diameter and the wall thickness at a plurality of closely spaced test points, which are circumferential and axial along the entire pipe length are distributed.

FIG. 2 also shows a transducer X3, which enables compensation of the echo reception times relative to the main pulses for all test transducers for changes in the temperature of the liquid coupling agent. As is known, the speed of propagation of ultrasound energy changes with the coupling agent with its temperature. For example, the speed of sound in water changes by 0.3% for every degree of temperature change. In order to obtain precise measurements of dimensionally small pipe wall thicknesses, temperature changes in the coupling agent must be taken into account. Therefore, the transducer X3 is arranged to direct a beam 20 of periodic ultrasonic energy pulses onto a target 22 and the pulse-echo propagation times through the coupling means 24 are processed to derive temperature dependent correction factors for use in the Processing of the times used for the pulse echo by the coupling means for the test transducers.

From Fig. 3 it can be seen that transducers X4 and X5 are positioned relative to the tube 12 so that they examine the tube wall in the circumferential direction for cracks such as defects, breaks, etc. Therefore, the transducer X4 is oriented so that it directs its highly focused beam 26 so that it hits test points on the outer tube surface at a suitable angle relative to the tangent, so that the ultrasound energy pulses between the inner and outer surfaces are zigzagged. Zack tracks are reflected. As is apparent from Fig. 3, the light emitted by the transducer X4 ultrasonic energy pulses follow zigzag paths in the circumferential direction through the pipe wall in the counterclockwise direction. On the other hand, the transducer X5 directs its beam 28 against the outer tube surface at the same angle so that the ultrasonic energy pulses coupled into the tube wall are reflected between the inner and outer surfaces along zigzag paths which run in the clockwise direction in the circumferential direction.

If there are no cracks, it practically spreads no ultrasound energy as echoes back to the transducers out. However, if there are cracks, the amplitudes are the echoes strongly depend on the crack orientation. If the larger dimension of a crack in a plane parallel to the zigzag paths, the echo amplitude may be too small to be detected. But if the bigger one Crack dimension in a plane essentially perpendicular to the zigzag tracks, enough ultrasound is Energy reflected by the crack to make it detectable Generate echo.

The test transducers X6 and X7 are directed so that they project their beams 30 and 32 at acute angles onto the pipe circumference so as to test the pipe wall in opposite longitudinal directions. As can be seen from FIG. 4, ultrasonic energy pulses from the transducer X6 are coupled into the tube wall and reflected in inner and outer tube surfaces in zigzag paths that run in the longitudinal direction to the right. Energy pulses from converter X7 check the wall cross section along zigzag paths that run lengthwise to the left. Again, no echoes are received if there are no cracks. Cracks that are primarily in planes substantially parallel to these longitudinally oriented zigzag paths reflect echoes of minimal amplitude, while cracks that are in planes substantially perpendicular to them reflect echoes with significant amplitudes. It is thus clear that, from the multidirectional tests of the transducers X4 to X7, all cracks in the pipe wall which are of impermissible size are detected by at least one of these transducers, regardless of the crack orientation. The echo amplitudes from a particular crack are processed to provide an indication of its size and type, and the pulse echo times are processed to indicate its location, such as whether it is in or near the inner or outer pipe surfaces. The size and location of a crack in the pipe wall are the main factors in deciding whether a pipe is acceptable or unacceptable.

In order to collect the echo test information obtained by the transducers X1 to X7 in a convenient, cost-effective manner, the signal processing circuit shown in FIG. 5 is used. As shown there, the transducers X1 to X7 are individually electrically excited in order to emit their main ultrasonic energy pulses through pulse generators or drivers D1 to D7. These drivers are triggered to electrically pulse their respective transducers by pulses P1 through P7, which are delivered in rapid succession by a synchronizer 40 , the operation of which is controlled by a clock 42 . The echo responses of these transducers are routed via separate RF channels CH1 to CH7, which contain separate attenuators A1 to A7, in order to separate input signals from a multiplexer 44 . These attenuators are set separately, programmable to take into account the different sensitivities of the transducers. The multiplexer is indexed by pulses P9, P11, P13, P15, P17 and P19 of synchronizer 40 to sequentially switch each of the RF channels to its single output line 46 in synchronism with the occurrence of RF echo signals.

The output signal of the multiplexer 44 is amplified by an amplifier 48 and immediately split into five channels CH8 to CH12, each of which contains an amplifier 50 . Channel CH8 delivers its echo signals to a CRT cathode ray tube oscilloscope for visual inspection. The echo signals on the channels CH9 to CH11 are fed to a time / digital converter TDC via separate discriminators 54 , 56 and 58 . The echo signals in channel CH12 are detected by a controllable peak detector GPD and discriminated by a separate discriminator 60 before being fed to the converter TDC. The output signal of the controllable peak value detector GPD is also fed via an analog / digital converter ADC to a computer 62 for analysis with regard to the crack size. In practice, the converter ADC can be included in the detector GPD. The converter TDC is provided with eight separate clocks, which can be started together and stopped independently for time stamp signals received by the discriminators. The times registered by each of these eight clocks are digitally encoded and fed to computer 62 at the end of each test cycle.

The discriminators 54 , 56 , 58 and 60 preferably form a constant ratio or a constant fraction. These discriminators generate a time stamp output signal to indicate when the wavefront of any echo signal reaches a value that is always a constant fraction of its peak amplitude. For this reason, the time control of the time stamp signals of the discriminator is relatively insensitive to echo signal amplitudes above a predetermined threshold.

In order to produce inspection windows that are precisely defined with respect to the time during which the echo signals on channels CH9 to CH12 are observed, discriminators 54 , 56 and 58 and controllable detector GPD are selectively controlled by a series of four Control generators G1 to G4. The generator G1 thus generates a control pulse in order to open a test window in the discriminator 54 . The control pulses provided by generator G2 open test windows in discriminator 56 , while generator G3 serves the same purpose with respect to discriminator 58 . The control pulses of the G4 generator open test windows in the GDP detector. The widths of these control pulses, which are supplied by the generators G1 to G4, and thus the duration of the individual test windows, are independently programmable by the computer 62 , as is indicated by the connection 66 . These control pulses are timed by pulses that are generated by the synchronizer 40 . Such as 5 can be seen from Fig., Trigger the Synchronisiererpulse P8, P10 and P12, the generators G1 to G3 in a corresponding manner to the control pulses to generate be entered during the Synchronisiererpulse P14, P16, P18 and P20 in an OR gate 68, to trigger the delivery of four successive control pulses by the generator G4. In practice, the control generator G4 can take the form of four separate control generators, which are correspondingly triggered by synchronizer pulses P14, P16, P18 and P20, the output pulses of the generator being routed via the OR gate to the peak value detector GPD. This enables the computer to individually program the widths of the control pulses and thus the widths of the test windows that are opened in the controllable peak detector.

In order to explain the operation of the echo signal processing circuit according to FIG. 5, reference is also made to FIG. 6, which shows the relative timings of the synchronizer pulses P1 to P22. Each test cycle is initiated by the delivery of the synchronizer pulse P1, which triggers the driver D1 for the pulse converter X1 so that it delivers its ultrasound main energy pulse. The pulse P1 is also supplied to start the eight clocks together in the converter TDC. The synchronizer pulses P2 to P7 follow in rapid succession, as can be seen in FIG. 6, in order to initiate the main pulses of the converters X2 to X7. The multiplexer 44 is initialized at the beginning of each test cycle, the main channel CHI being connected through to its output line 46 . After a suitable delay, which is timed with respect to the return of test echoes to the converter X1, the synchronizer pulse P8 triggers the generator G1 to open the test window of the discriminator 54 . This state is shown in Fig. 6 by the timing line G1. At the end of the control pulse G1 to close this test window, the synchronization pulse P9 indexes the multiplexer 44 to the channel CH2, and then the synchronizer pulse P10 triggers the generator G2 to deliver its control pulse to open a test window in the discriminator 56 . After this window is closed, the synchronizer pulse P11 switches the multiplexer to the channel CH3, and the synchronizer pulse P12 triggers the generator G3 to open a test window in the discriminator 58 . Synchronizer pulses P13 to P20 follow one another to step the multiplexer through channels CH4 to CH7 and trigger generator G4 to open four successive test windows in the controllable peak detector GPD, as shown in FIG. 6.

The synchronizer pulse P21 then signals the end of a test cycle and is used to reset to zero those clocks in the converter TDC that have not been stopped by output signals from the discriminators 54 , 56 , 58 and 60 . This pulse P21 is also used by the converter to signal the computer 62 to accept digital data over a cable 70 indicating the times used up registered by each of the eight clocks. This data reading takes place during the interval RD, which is shown in FIG. 6. Finally, the synchronizer pulse P22 puts the multiplexer 44 back on the channel CHI and initiates another test cycle, which is marked by the delivery of the pulse G1. The test cycles are repeated automatically in order to thoroughly test a pipe over its entire length, both in terms of its dimensions and for cracks.

The operation of the converter TDC can best be understood from FIG. 7. As already stated, the synchronizer pulse P1 starts each test cycle by triggering the driver D1 to initiate the main pulse of the converter X1. Timeline G1 in FIG. 7 represents the test window that was opened in discriminator 54 to detect echoes received by this transducer in response to its main ultrasound energy pulse. From a joint consideration of FIGS. 2 and 7 shows that the first signal 72 that occurs in this test window, the echo, which has been reflected to the transducer X1 from the outer surface of the tube 12, while the signal 73 the echo which has been reflected from the boundary of the inner tube surface to its open bore which is filled with air or a coupling fluid. The third signal 74 that occurs in the test window G1 is the echo of ultrasonic energy that is reflected from the boundary of the outer pipe surface and the coupling means 24 to the inner pipe surface boundary and then back to the transducer X1. In response to the echo signal 72 , the discriminator 54 outputs a first time stamp signal to the converter TDC (see FIG. 5), which stops one of its eight clocks at the elapsed time T1 measured by the synchronizer pulse P1. The discriminator 54 outputs a second time stamp signal in response to the echo signal 73 to stop a second detector clock at the used time P2, and it outputs a third time stamp signal to stop a third clock at the used time T3, like all of these is shown in Fig. 7.

The test window of generator G2 is then opened in discriminator 56 to detect outer tube surface echo signal 75 , inner tube surface echo signal 76 and reflected echo signal 77 received by transducer X2. The time stamp signals of the discriminators, which are supplied in response to the echo signals 75 to 77 , stop the next three converter clocks at the times T4, T5 and T6 used up.

From these six clock counts, computer 62 has enough data to calculate the pipe dimensions at the particular diametrically opposed test points that have been examined by transducers X1 and X2 during each test cycle. That is, the subtraction of the distances, which are calculated from the times T1 and T4 used, from the distance D, which separates the transducers X1 and X2, gives the outer diameter of the tube. The pipe wall thickness at the test point of the converter X1 is calculated from the difference between the times T2 and T3 used. Similarly, the wall thickness at the test site of converter X2 is calculated by the computer from the difference between the times T5 and T6 used. The inside diameter of the tube can then be calculated from the outside diameter determined and two wall thickness determinations.

In order to obtain precise measurements of the pipe dimensions, it has been pointed out that temperature changes in the coupling means must be compensated for. Therefore, the G3 test window in the discriminator 58 is opened to detect the echo signal 78 which is reflected from the target 22 to the transducer X3 (see FIG. 2). In response to this signal, discriminator 58 outputs a timestamp signal that stops the seventh clock in converter TDC at time T7 from synchronizer pulse P1, as shown in FIG. 7. The computer 62 uses this thus determined used time T7 to correct the used times T1 to T6 and thus to correct the pipe dimension calculations for changes in the sound propagation speed in the fluid coupling means 24 due to changes in its temperature.

7 is further apparent from FIG. That four successive scan window to be opened in response to the Synchronisiererpulse P14, P16, P18 and P20 in the controllable peak detector GPD, are shown in G4A, G4B, G4C G4D and. These windows are timed so that they open immediately after receiving echo signals 80 to 83 by transducers X4 to X7 from the outer tube surface (see FIGS. 3 and 4). If there are no cracks, no echoes are received in these test windows, as is indicated in the case of windows G4A, G4B and G4D. However, if the ultrasonic energy propagating through the tube wall in its zigzag path encounters a crack, a crack echo is received in one of these test windows, such as a crack signal 84 received by transducer X6 and in the window G4C is arranged. The peak-to-peak value of this crack signal is digitized in the ADC ADC and enters the computer 62 as an indication of the crack size. In addition, in response to this crack signal 84, the discriminator 60 emits a time stamp signal in order to stop the eighth clock generator in the converter TDC after the elapsed time T8. This time value is processed by the computer in order to determine which crack test transducer X4 to X7 has received the crack echo, in order to also determine the crack position within the tube wall cross section. If no crack signal is detected during a test cycle, the eighth converter clock is set to zero by the synchronization pulse 21 and the resulting zero count is interpreted by the computer as an absence of a crack.

It should be noted that the converter TDC quickly with additional clocks could be equipped to the remote way to handle that Crack signals in several or even all four test windows G4A to G4D detected during a single test cycle will. Alternatively, the four crack test transducers  Echo signals over separate signal channels instead of through the single channel CH12 can be handled, each Crack signal channel with a separate clock in the converter TDC is connected.

It is also possible that the axial and rotational movements of the pipe are tracked through the test station so that the computer precisely locates the test points on the pipe can, each dimension and crack test by transducer X1 and X2 and X4 to X7 during each test cycle is brought about. Thus, the pipe locations with Out of tolerance dimensions and cracks identified by for a subsequent visual inspection Quality assurance staff.

Claims (9)

1. Device for non-destructive crack and dimension testing of thin-walled tubular elements, comprising:

• - A. First and second ultrasonic transducers (X1, X2) for dimension testing of a tubular element ( 12 ) along helical scanning paths, the first and second transducers (X1, X2) each at a known distance (D) from one another in a diametrically opposite position are arranged to the tubular element ( 12 ) and generate a sequence of first, second and third echo signals as a response to ultrasonic energy pulses, which are reflected from the outer surface, the inner surface and again from the inner surface of the tubular element ( 12 ) ,
• B. a number of ultrasonic transducers (X4, X5) for crack testing of the tubular element ( 12 ) along helical scanning paths,
• - C. Driver (D1-D7) for electrically exciting the first and second transducers (X1, X2) and the number of ultrasound transducers (X4, X5) in order to emit ultrasound energy pulses, the transducers all being provided by a fluid coupling means ( 24 ) are directed towards the tubular element ( 12 ),
• D. a separate RF channel assigned to each transducer for forwarding the received echo signals to a multiplexer ( 44 ) which has a single output,
• - E. an amplifier ( 48 ) whose input is connected to the output of the multiplexer ( 44 ),
• - F. a signal divider with an input connected to the amplifier output and several outputs,
• G. separate signal detector channels (CH8-CH12), which are each connected to a signal divider output and each have a discriminator ( 54-60 ) for generating a time stamp signal as a function of an echo signal, where at least one of the detector channels (CH8-CH12) Has peak detector (GPD) for generating a signal indicating the peak-to-peak amplitude of an echo signal,
• - H. separate time control means (TDC), each of which is connected to a detector channel, the first, second and third echo signals of the first converter (X1) being detected by one of the discriminators in a first detector channel and first, second and third time stamp signals are generated, and the first, second and third echo signals of the second converter (X2) are detected by one of the discriminators in a second detector channel and fourth, fifth and sixth time stamp signals are generated, the time control means (TDC) having a plurality of clocks which are common be started at the beginning of a test cycle and stopped separately by the time stamp signals, such that data relating to the time consumed are supplied from which a computer ( 62 ) can calculate dimension data for the tubular element ( 12 ), and
• - I. Synchronizer means (G1-G4) for generating a sequence of timing pulses for synchronization
• 1. the electrical excitation of all transducers (X) by the drivers (D1-D7),
• 2. The operation of the multiplexer ( 44 ) when successively connecting each of its inputs to its single output and
• 3. the time control means (TDC) such that the times used up between the delivery of the ultrasonic energy test pulses and the reception of the corresponding ultrasonic energy pulse echoes are displayed separately by each transducer.

2. Device according to claim 1, characterized in that the timing pulses control a device for generating test windows in the discriminators ( 54-60 ) and the peak value detector (GPD). 3. Device according to claim 2, characterized in that the computer ( 62 ) is connected so that it receives the signals indicating a peak amplitude from the peak value detector (GPD). 4. Device according to claim 1, characterized in that the time control means (TDC) a time / digital converter have the separate information of the used Passes times to the computer. 5. Device according to claim 4, characterized in that the echo signals through the transducers (X4, X5) of the second Number generated by a single peak detector (GPD) in the one detector channel (CH12) successively can be detected. 6. Device according to claim 5, characterized in that the number of transducers (X4, X5, X6, X7) are arranged relative to the tubular element ( 12 ) such that the sen wall is internally testable by ultrasonic energy pulses, which propagate in opposite longitudinal and circumferential directions, with an echo signal detected by a transducer of the number in response to ultrasonic energy pulses reflected from a crack by the peak detector to indicate the signals indicating peak-to-peak amplitude and that a discriminator generates a seventh time stamp signal in a third detector channel, a separate clock in the time control device (TDC) being stopped by the seventh time stamp signal, so that a value for the time spent is determined, from which the computer ( 62 ) determines the crack position can. 7. Device according to claim 6, characterized in that a compensating converter (X3) is provided for generating a reference echo signal, wherein one of the discriminators in a fourth detector channel generates an eighth time mark signal in response to the reference echo signal for stopping a separate clock in the timing device (TDC) to provide data on the time spent indicating the rate of propagation of ultrasonic energy in the fluid coupling medium, from which the computer ( 62 ) can calculate a correction factor, the changes in temperature of the coupling agent compensated. 8. Device according to claim 6, characterized in that the discriminators ( 54 , 60 ) have a constant ratio or a constant fraction. 9. Device according to claim 8, characterized in that an additional signal detector channel (CH8) to conduct Echo signals to a cathode ray tube display (CRT) is provided.